cod.py

import tweepy
import openai
import matplotlib.pyplot as plt
import pandas as pd
import time

# -----------------------------
# Configuration (Replace with your API keys)
# -----------------------------
TWITTER_API_KEY = "YOUR_TWITTER_API_KEY"
TWITTER_API_SECRET = "YOUR_TWITTER_API_SECRET"
TWITTER_ACCESS_TOKEN = "YOUR_TWITTER_ACCESS_TOKEN"
TWITTER_ACCESS_TOKEN_SECRET = "YOUR_TWITTER_ACCESS_TOKEN_SECRET"
OPENAI_API_KEY = "YOUR_OPENAI_API_KEY"

# -----------------------------
# Twitter API Setup
# -----------------------------
auth = tweepy.OAuth1UserHandler(
    TWITTER_API_KEY, TWITTER_API_SECRET,
    TWITTER_ACCESS_TOKEN, TWITTER_ACCESS_TOKEN_SECRET
)
twitter_api = tweepy.API(auth)

# -----------------------------
# Data Collection Function
# -----------------------------
def get_tweets(query, count=50):
    """
    Fetches tweets based on a query.
    """
    try:
        tweets = twitter_api.search_tweets(q=query, count=count, lang="en", tweet_mode="extended")
        tweet_texts = [tweet.full_text for tweet in tweets]
        return tweet_texts
    except Exception as e:
        print("Error fetching tweets:", e)
        return []

# -----------------------------
# Sentiment Analysis via OpenAI
# -----------------------------
def analyze_sentiment(text):
    """
    Uses OpenAI's API to analyze sentiment.
    Returns sentiment as a string (e.g., Positive, Negative, Neutral).
    """
    openai.api_key = OPENAI_API_KEY
    try:
        # Define a simple prompt for sentiment analysis
        prompt = f"Analyze the sentiment of the following text and return only the sentiment (Positive, Negative, or Neutral):\n\n{text}"
        response = openai.ChatCompletion.create(
            model="gpt-3.5-turbo",  # or gpt-4 if available
            messages=[{"role": "user", "content": prompt}],
            max_tokens=10,
            n=1,
            stop=None,
            temperature=0.5
        )
        sentiment = response.choices[0].message["content"].strip()
        return sentiment
    except Exception as e:
        print("Error analyzing sentiment:", e)
        return "Unknown"

# -----------------------------
# Visualization Function
# -----------------------------
def visualize_results(results):
    """
    Visualizes the sentiment results in a bar chart.
    """
    df = pd.DataFrame(results, columns=["Text", "Sentiment"])
    sentiment_counts = df["Sentiment"].value_counts()
    
    plt.figure(figsize=(8, 6))
    sentiment_counts.plot(kind="bar", color=['green', 'red', 'blue'])
    plt.xlabel("Sentiment")
    plt.ylabel("Number of Tweets")
    plt.title("Sentiment Analysis of Tweets")
    plt.xticks(rotation=0)
    plt.tight_layout()
    plt.show()

# -----------------------------
# Main Execution
# -----------------------------
if __name__ == "__main__":
    query = "local transportation issues"  # Customize your search query
    print("Fetching tweets for query:", query)
    tweets = get_tweets(query, count=50)
    
    if not tweets:
        print("No tweets found or an error occurred.")
    else:
        results = []
        print("Analyzing sentiment of fetched tweets...")
        for tweet in tweets:
            sentiment = analyze_sentiment(tweet)
            results.append((tweet, sentiment))
            time.sleep(1)  # Pause to respect API rate limits
        
        # Show a sample of results
        for i, (text, sentiment) in enumerate(results[:3]):
            print(f"Tweet {i+1}: {sentiment} -> {text[:80]}...")
        
        # Visualize the sentiment distribution
        visualize_results(results)